For the best experience on desktop, install the Chrome extension to track your reading on news.ycombinator.com
Hacker Newsnew | past | comments | ask | show | jobs | submit | history | mlajtos's commentsregister

This is the first part of the "New Kind of Paper" series that deals with concept of "handwritten calculator" – you scribble "1+2" and the system responds with "3". Think "Apple Math Notes", but 3 years before its debut. And with vectors, matrices, tensors, etc.

There are 5 parts so far:

1. https://mlajtos.mu/posts/new-kind-of-paper – overview of the concept

2. https://mlajtos.mu/posts/new-kind-of-paper-2 – intro to Fluent notation

3. https://mlajtos.mu/posts/new-kind-of-paper-3 – inspirational prior inventions

4. https://mlajtos.mu/posts/new-kind-of-paper-4 - how to think efficiently (including longer demo)

5. https://mlajtos.mu/posts/new-kind-of-paper-5 – review of Apple Math Notes and suggestions for improvements

Keyboard version of Fluent is available here: https://mlajtos.github.io/fluent/?code=RG9jdW1lbnRhdGlvbg


Fluent – 4K lines – including parser, interpreter, standard library, IDE, UI, docs, examples. Will grow though.

https://github.com/mlajtos/fluent/blob/main/client.tsx


I'll add it! Thanks.

EDIT: Actually, it's not quite "ML-family" enough for this post. But it is a remarkably cool project! :)


If you're accepting additions here is a fun one some friends and I did as experiment at the Topos Institute: https://github.com/ToposInstitute/polytt

And here is a set of single file lambda calculus implementations with a variety of extensions: https://github.com/solomon-b/lambda-calculus-hs

`polytt` is kind of an ended experiment but that lambda calculus repo i plan to extend in the near future.


> Actually, it's not quite "ML-family" enough for this post. But it is a remarkably cool project! :)

Fair. Thank you :)


Shake your Mac to undo would be consistent with iPhone, iPad, Vision Pro. Yes, you shake your head to undo when wearing AVP.


This is a milder version of “Shake to Reboot” which is standard for all Etch-a-Sketch models


Does it support API key access or only Claude.ai subscription?


Worth noting that "API key access" vs "subscription" has significant cost implications for heavy users.

Claude.ai Pro is $20/month flat. But if you're doing serious agent-assisted coding (multi-file refactors, iterative debugging loops), you can blow through $50-100/day in API costs.

The math changes depending on usage patterns. Subscription makes sense for interactive coding sessions. API keys make sense if you're batch processing or running agents autonomously overnight.


I am doing interactive coding sessions via API. I don't want to see a message that I am over limit to use the best model there is.


Both


I was hoping for 6DoF sub-mm realtime tracking. My disappointment is immeasurable and my day is ruined.


Would be great, but that, world-wide, for millions of objects, probably is a case of putting your expectations too high.

Has that ever been demonstrated for a single object, even if allowing the object to be a thousand times as large as this?


Just have an unique ipv6 address on every object. Easy.


why do you need that?


Pure visual object tracking in visionOS is considerable laggy (even with increased detection rate). Natively tracked peripherals (Logitech Muse, PSVR2 controllers) are super responsive, but are designed for hands and are too specialized. There is a place for generic 6DoF tracking device that can be attached to any object you want to track. This could be tiny IR LED array if you want to track it inside the field of view, but when you need precise position outside of your FoV, your options are limited.


You are exactly right that the original vision of APL was a handwritten, non-ambiguous, consistent, and executable math notation. This was never accomplished.

In 2021, I made a prototype calculator designed for iPad and Pencil and wrote four essays called "New Kind of Paper" describing this concept. The video demonstration of the prototype [Demo] sums up pretty well how it might be used.

In 2024, Apple released Math Notes, which implements this concept for standard math notation. My "review" of it is at [MathNotes]. In short, it is currently a preview of a damn great tool, but its future depends on notation and expressivity.

APL is a language from 2066, created in 1966. While its semantics make more sense now due to machine learning, syntactically it remains alien to most people. Backus' FP/FL and Iverson's J are more approachable to current programmers, but still not there. In New Kind of Paper, I created a tiny language called Fluent, which is nowhere near the ambition of any language mentioned, but it is intentionally designed to be handwritten. A week ago, I open-sourced it. [Fluent]

There is plenty of work left to do, but it is a good start.

[NKoP]: https://mlajtos.mu/posts/new-kind-of-paper

[MathNotes]: https://mlajtos.mu/posts/new-kind-of-paper-5

[Fluent]: https://news.ycombinator.com/item?id=46649223

[Demo]: https://youtu.be/y5Tpp_y2TBk

Thanks for coming to my TED talk.


yes


This is absolutely insane in the most positive way.


I have a project using tfjs and jax-js is very exciting alternative. However during porting I struggle a lot with `.ref` and `.dispose()` API. Coming from tfjs where you garbage collect with `tf.tidy(() => { ... })`, API in jax-js seems very low-level and error-prone. Is that something that can be improved or is it inherent to how jax-js works?

Would `using`[0] help here?

[0]: https://developer.mozilla.org/en-US/docs/Web/JavaScript/Refe...


I don’t think tf.tidy() is a sound API under jvp/grad transformations, also it prevents you from using async which makes it incompatible with GPU backends (or blocks the page), a pretty big issue. https://github.com/tensorflow/tfjs/issues/5468

Thanks for the feedback though, just explaining how we arrived at this API. I hope you’d at least try it out — hopefully you will see when developing that the refs are more flexible than alternatives.


I'll grind jax-js more and see if refs become invisible then. Thanks for a great project!


This problem is more suitable for tiny LLM.


In my experience, LLMs are terrific for most translation tasks, but you still need a way to encode the data (rules for genders, cardinalities, ordinalities, ...) for storage on disk/database/etc. for 1. performance and 2. consistency/durability. So LLMs are a big part of the solution, but not the whole picture.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search:

HN For You